F1 Layout

Map indicates the layout of the F1 generation resulting from a cross between EC201 and EC103 parents. Column 1 is approximately lengthways facing north.

“Map of F1 Trees”

“Map of F1 Trees”

Height Diagram

Diagram indicates the areas of leaf collection regarding height. Each Tree had 10 leaves collected, 3 from Low, 4 from Mid and 3 from High. The first leaf sampled was measured twice for replication comparison.

“Diagram of leaf collection levels”

“Diagram of leaf collection levels”

Import and Arrange Data

“Full.xlsx” contains measurement information from sampling with the Dualex (https://www.force-a.com/en/capteurs-optiques-optical-sensors/dualex-scientific-chlorophyll-meter/), including;

It also contains information about the block position, the leaf height information, and presense or absence of flowering

Sheet “Dup” contains only the samples that were replicated.

# Import Data Measures
Data <-read.xlsx("Full.xlsx", sheetName ="Full")
head(Data)
##   Collection.Day Allocation Block Column Row group Group.ID Tree.ID Rep.
## 1              4         CG     0      0   0    20       OG      CG    N
## 2              4         CG     0      0   0    20       OG      CG    N
## 3              4         CG     0      0   0    20       OG      CG    N
## 4              4         CG     0      0   0    20       OG      CG    N
## 5              4         CG     0      0   0    20       OG      CG    N
## 6              4         CG     0      0   0    20       OG      CG    N
##   measure Height Flower    Chl  Flav  Anth   NBI
## 1       9      H      Y 28.220 2.418 0.205 11.67
## 2       3      L      Y 27.958 2.298 0.691 12.17
## 3      11      H      Y 33.727 2.458 0.527 13.72
## 4       4      L      Y 25.938 1.758 0.172 14.76
## 5      10      H      Y 36.205 2.283 0.270 15.86
## 6       6      M      Y 34.332 2.115 0.150 16.23
Data$Column = as.factor(Data$Column)
Data$Row = as.factor(Data$Row)

# Import Replicate Data
Dup <-read.xlsx("Full.xlsx", sheetName ="Dup")
head(Dup)
##   Collection.Day group Group.ID Tree.ID Rep. measure Height   Chl  Flav
## 1            2.0    23       60   IN4DV   Y1       1      L 1.916 2.363
## 2            1.5     3        1   IN4BT   Y1       1      L 3.124 2.300
## 3            2.0    17       54   IN4DL   Y2       2      L 3.414 1.826
## 4            1.5    32       28   IN4CP   Y2       2      L 4.097 1.943
## 5            2.0     9       46   IN4DC   Y1       1      L 4.909 1.928
## 6            1.5     5        3   IN4BW   Y1       1      L 4.924 1.848
##    Anth  NBI
## 1 0.174 0.81
## 2 0.191 1.36
## 3 0.112 1.87
## 4 0.051 2.11
## 5 0.103 2.55
## 6 0.165 2.66
#Isolate Crimson Glory Outgroup
CG = Data[c(1:11),]
head(CG)
##   Collection.Day Allocation Block Column Row group Group.ID Tree.ID Rep.
## 1              4         CG     0      0   0    20       OG      CG    N
## 2              4         CG     0      0   0    20       OG      CG    N
## 3              4         CG     0      0   0    20       OG      CG    N
## 4              4         CG     0      0   0    20       OG      CG    N
## 5              4         CG     0      0   0    20       OG      CG    N
## 6              4         CG     0      0   0    20       OG      CG    N
##   measure Height Flower    Chl  Flav  Anth   NBI
## 1       9      H      Y 28.220 2.418 0.205 11.67
## 2       3      L      Y 27.958 2.298 0.691 12.17
## 3      11      H      Y 33.727 2.458 0.527 13.72
## 4       4      L      Y 25.938 1.758 0.172 14.76
## 5      10      H      Y 36.205 2.283 0.270 15.86
## 6       6      M      Y 34.332 2.115 0.150 16.23
#Isolate East Cape 201 Parent
EC201 = Data[c(12:21),]
head(EC201)
##    Collection.Day Allocation Block Column Row group Group.ID Tree.ID Rep.
## 12              4      EC201     0      0   0    18       P1   EC201    N
## 13              4      EC201     0      0   0    18       P1   EC201    N
## 14              4      EC201     0      0   0    18       P1   EC201    N
## 15              4      EC201     0      0   0    18       P1   EC201    N
## 16              4      EC201     0      0   0    18       P1   EC201    N
## 17              4      EC201     0      0   0    18       P1   EC201    N
##    measure Height Flower    Chl  Flav  Anth   NBI
## 12       9      H      Y 30.722 2.138 0.156 14.37
## 13       6      M      Y 45.272 1.861 0.075 24.33
## 14      10      H      Y 46.174 1.868 0.052 24.72
## 15       8      M      Y 57.480 1.897 0.037 30.30
## 16       5      M      Y 57.752 1.681 0.125 34.35
## 17       3      L      Y 54.927 1.592 0.061 34.50
#Isolate East Cape 103 Parent
EC103 = Data[c(22:33),]
head(EC103)
##    Collection.Day Allocation Block Column Row group Group.ID Tree.ID Rep.
## 22              4      EC103     0      0   0    19       P2   EC103    N
## 23              4      EC103     0      0   0    19       P2   EC103    N
## 24              4      EC103     0      0   0    19       P2   EC103    N
## 25              4      EC103     0      0   0    19       P2   EC103    N
## 26              4      EC103     0      0   0    19       P2   EC103    N
## 27              4      EC103     0      0   0    19       P2   EC103    N
##    measure Height Flower    Chl  Flav  Anth   NBI
## 22       8      M      Y 31.471 1.324 0.018 23.77
## 23      11      H      Y 57.349 2.111 0.053 27.17
## 24       4      L      Y 43.949 1.478 0.004 29.73
## 25      12      H      Y 56.779 1.846 0.108 30.75
## 26       9      M      Y 55.916 1.746 0.055 32.03
## 27      10      H      Y 57.809 1.746 0.054 33.11
#Isolate Offspring from the Parental Cross
F1 = Data[c(34:1825),]
head(F1)
##    Collection.Day Allocation Block Column Row group Group.ID Tree.ID Rep.
## 34              1         F1     1      1   1    15        1   IN4BT    N
## 35              1         F1     1      1   1    15        1   IN4BT    N
## 36              1         F1     1      1   1    15        1   IN4BT    N
## 37              1         F1     1      1   1    15        1   IN4BT    N
## 38              1         F1     1      1   1    15        1   IN4BT    N
## 39              1         F1     1      1   1    15        1   IN4BT   Y1
##    measure Height Flower    Chl  Flav  Anth   NBI
## 34       8      M      N  3.268 1.958 0.143  1.67
## 35       2      L      N 23.326 2.581 0.065  9.04
## 36      11      M      N 24.682 2.265 0.112 10.90
## 37       7      M      N 27.068 2.229 0.084 12.14
## 38      14      H      N 27.459 2.135 0.138 12.86
## 39       3      L      N 34.592 2.583 0.048 13.39

Replicate Analysis

Replicates were taken by measuring a single leaf sample from each tree twice, in order to establish consistency and reliability of measurements with the Dualex.

Replicate Data Overview
#Anthocyanin
RepAnth = summary(Dup$Anth)

#Chlorophyll
RepChl = summary(Dup$Chl)

#Flavanoid
RepFlav = summary(Dup$Flav)

#NBI
RepNBI =summary(Dup$NBI)

as.table(rbind(RepAnth,RepChl,RepFlav,RepNBI))
##                Min.     1st Qu.      Median        Mean     3rd Qu.
## RepAnth  0.00100000  0.06500000  0.09850000  0.09799367  0.12525000
## RepChl   1.91600000 25.82475000 36.02700000 35.76269937 46.61000000
## RepFlav  1.05600000  1.76775000  1.99400000  1.96720570  2.19725000
## RepNBI   0.81000000 12.70250000 18.19000000 18.73398734 24.45750000
##                Max.
## RepAnth  0.25400000
## RepChl  59.67400000
## RepFlav  2.71400000
## RepNBI  43.49000000
Replicate Group means
Dup2 = group_by(Dup,Rep.)
Dup2 = summarise(Dup2, Anth = mean(Anth, na.rm = T),Chl = mean(Chl, na.rm = T), Flav = mean(Flav,na.rm = T), NBI = mean(NBI, na.rm = T))
head(Dup2)
## # A tibble: 2 x 5
##   Rep.    Anth   Chl  Flav   NBI
##   <fct>  <dbl> <dbl> <dbl> <dbl>
## 1 Y1    0.0983  34.9  1.98  18.2
## 2 Y2    0.0977  36.6  1.95  19.2
Replicate Plots
#Anthocyanin Replicate Plots
plot1<- ggplot(Dup) + aes(x = Rep., y = Anth ) + geom_boxplot() + ylab("Anthocyanin Measure") + xlab("Replicate Number") + ggtitle("Anthocyanin of Replicate Samples") +stat_summary(fun.y=mean, colour="darkred", geom="point", shape=18, size=3);plot1

plot11<-ggplot(Dup) +aes(x = Tree.ID, y = Anth) + geom_point() +geom_hline(yintercept = mean(Dup$Anth))+ ylab("Anthocyanin Measure") + xlab("Tree ID") + ggtitle("Anthocyanin of Replicate Samples")
plot12 <-ggplot(Dup) +aes(Anth) +geom_histogram() + stat_bin(binwidth = 0.001) + ylab("Frequency") + xlab("Anthocyanin Measure") + ggtitle("Anthocyanin of Replicate Samples")
grid.arrange(plot11,plot12)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Chlorophyll Replicate Plots
plot2<- ggplot(Dup) + aes(x = Rep., y = Chl) + geom_boxplot()+ ylab("Chlorophyll Measure") + xlab("Replicate Number") + ggtitle("Chlorophyll of Replicate Samples")+stat_summary(fun.y=mean, colour="darkred", geom="point", shape=18, size=3,show_guide = FALSE);plot2
## Warning: `show_guide` has been deprecated. Please use `show.legend`
## instead.

plot21 <-ggplot(Dup) +aes(x = Tree.ID, y = Chl) + geom_point() +geom_hline(yintercept = mean(Dup$Chl))+ ylab("Chlorophyll Measure") + xlab("Tree ID") + ggtitle("Chlorophyll of Replicate Samples")
plot22 <-ggplot(Dup) +aes(Chl) +geom_histogram() + stat_bin(binwidth = 2) + ylab("Frequency") + xlab("Chlorophyll Measure") + ggtitle("Chlorophyll of Replicate Samples")
grid.arrange(plot21,plot22)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Flavonol Replicate Plots
plot3<- ggplot(Dup) + aes(x = Rep., y = Flav ) + geom_boxplot()+ ylab("Flavonol Measure") + xlab("Replicate Number") + ggtitle("Flavonol of Replicate Samples")+stat_summary(fun.y=mean, colour="darkred", geom="point", shape=18, size=3);plot3

plot31 <-ggplot(Dup) +aes(x = Tree.ID, y = Flav) + geom_point() +geom_hline(yintercept = mean(Dup$Flav))+ ylab("Flavonol Measure") + xlab("Tree ID") + ggtitle("Flavonol of Replicate Samples")
plot32 <-ggplot(Dup) +aes(Flav) +geom_histogram() + stat_bin(binwidth = 0.001)+ ylab("Frequency") + xlab("Flavonol Measure") + ggtitle("Flavonol of Replicate Samples")
grid.arrange(plot31,plot32)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#NBI Replicate Plots
plot4<- ggplot(Dup) + aes(x = Rep., y = NBI ) + geom_boxplot()+ ylab("NBI Measure") + xlab("Replicate Number") + ggtitle("NBI of Replicate Samples")+stat_summary(fun.y=mean, colour="darkred", geom="point", shape=18, size=3);plot4

plot41 <-ggplot(Dup) +aes(x = Tree.ID, y = NBI) + geom_point() +geom_hline(yintercept = mean(Dup$NBI)) + ylab("NBI Measure") + xlab("Tree ID") + ggtitle("NBI of Replicate Samples")
plot42 <-ggplot(Dup) +aes(NBI) +geom_histogram() + stat_bin(binwidth = 2)+ ylab("Frequency") + xlab("NBI Measure") + ggtitle("NBI of Replicate Samples")
grid.arrange(plot41,plot42)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

There is no statistically significant differences between the two groups of measurements, this is a good sign indicative of the accuracy of the Dualex.

ANOVA of Replicates
#Chlorophyll Anova
mod <-lm(Dup$Chl ~ Dup$Rep.)
anova(mod)
## Analysis of Variance Table
## 
## Response: Dup$Chl
##            Df Sum Sq Mean Sq F value Pr(>F)
## Dup$Rep.    1    218  218.33   1.033 0.3102
## Residuals 314  66366  211.36
#Flavanol Anova
mod2 <-lm(Dup$Flav ~ Dup$Rep.)
anova(mod2)
## Analysis of Variance Table
## 
## Response: Dup$Flav
##            Df  Sum Sq  Mean Sq F value Pr(>F)
## Dup$Rep.    1  0.0521 0.052129  0.6506 0.4205
## Residuals 314 25.1596 0.080126
#Anthocyanin Anova
mod3 <-lm(Dup$Anth ~ Dup$Rep.)
anova(mod3)
## Analysis of Variance Table
## 
## Response: Dup$Anth
##            Df  Sum Sq   Mean Sq F value Pr(>F)
## Dup$Rep.    1 0.00003 0.0000304  0.0151 0.9023
## Residuals 314 0.63303 0.0020160
#NBI Anova
mod4 <-lm(Dup$NBI ~ Dup$Rep.)
anova(mod4)
## Analysis of Variance Table
## 
## Response: Dup$NBI
##            Df  Sum Sq Mean Sq F value Pr(>F)
## Dup$Rep.    1    82.5  82.530   1.115 0.2918
## Residuals 314 23241.0  74.016

The absence of statistically significant results indicates that our replicates are likely to be consistent.

Allocation Analysis

Allocation refers to which group measurements were taken from, i.e. A Parental Tree (EC103 or EC201), Outgroup Tree (CG), Parental Offspring (F1)

#Anthocyanin Mean
AllAnth = summary(Data$Anth)
#Chlorophyll Mean
AllChl = summary(Data$Chl)
#Flavanol Mean
AllFlav = summary(Data$Flav)
#NBI Mean
AllNBI = summary(Data$NBI)

as.table(rbind(AllAnth,AllChl,AllFlav,AllNBI))
##                Min.     1st Qu.      Median        Mean     3rd Qu.
## AllAnth  0.00100000  0.06800000  0.09600000  0.09772877  0.12400000
## AllChl   0.13000000 24.10200000 36.21000000 35.21443342 46.82300000
## AllFlav  1.05600000  1.80600000  1.97800000  1.96440493  2.12800000
## AllNBI   0.07000000 12.14000000 18.64000000 18.34807123 24.44000000
##                Max.
## AllAnth  0.69100000
## AllChl  59.90500000
## AllFlav  2.86100000
## AllNBI  49.17000000
Allocation Group Means
Data2 = group_by(Data, Allocation)
Data2 = summarise(Data2, Anth = mean(Anth, na.rm = T),Chl = mean(Chl, na.rm = T), Flav = mean(Flav, na.rm = T), NBI = mean(NBI, na.rm = T))
head(Data2)
## # A tibble: 4 x 5
##   Allocation   Anth   Chl  Flav   NBI
##   <fct>       <dbl> <dbl> <dbl> <dbl>
## 1 CG         0.258   36.0  2.17  16.7
## 2 EC103      0.0588  52.6  1.60  33.1
## 3 EC201      0.0779  52.6  1.71  31.7
## 4 F1         0.0971  35.0  1.97  18.2
CGAnth = summary(CG$Anth)
F1Anth = summary(F1$Anth)
Anth201 = summary(EC201$Anth)
Anth103 = summary(EC103$Anth)

CGChl = summary(CG$Chl)
F1Chl = summary(F1$Chl)
Chl201 = summary(EC201$Chl)
Chl103 = summary(EC103$Chl)


CGFlav = summary(CG$Flav)
F1Flav = summary(F1$Flav)
Flav201 = summary(EC201$Flav)
Flav103 = summary(EC103$Flav)

CGNBI = summary(CG$NBI)
F1NBI = summary(F1$NBI)
NBI201 = summary(EC201$NBI)
NBI103 = summary(EC103$NBI)

as.table(rbind(CGAnth,F1Anth,Anth201,Anth103,CGChl,F1Chl,Chl201,Chl103,CGFlav,F1Flav,Flav201,Flav103,CGNBI,F1NBI,NBI201,NBI103))
##                Min.     1st Qu.      Median        Mean     3rd Qu.
## CGAnth   0.10700000  0.16100000  0.20100000  0.25763636  0.24250000
## F1Anth   0.00100000  0.06800000  0.09650000  0.09711830  0.12400000
## Anth201  0.03700000  0.05425000  0.06850000  0.07790000  0.08400000
## Anth103  0.00400000  0.03725000  0.05350000  0.05883333  0.08925000
## CGChl   25.93800000 30.97350000 35.70700000 36.02836364 40.68400000
## F1Chl    0.13000000 23.77200000 36.09900000 34.99601172 46.63200000
## Chl201  30.72200000 48.36225000 56.52100000 52.60670000 58.57325000
## Chl103  31.47100000 52.16025000 55.50900000 52.59241667 57.46400000
## CGFlav   1.75800000  2.06200000  2.11500000  2.16945455  2.29050000
## F1Flav   1.05600000  1.80700000  1.97800000  1.96700279  2.12800000
## Flav201  1.45400000  1.56350000  1.63850000  1.71080000  1.86625000
## Flav103  1.29300000  1.46400000  1.55450000  1.59983333  1.74600000
## CGNBI   11.67000000 14.24000000 16.23000000 16.73909091 18.93000000
## F1NBI    0.07000000 12.04750000 18.52500000 18.18443080 24.31500000
## NBI201  14.37000000 26.11500000 34.42500000 31.68800000 36.59000000
## NBI103  23.77000000 30.49500000 33.40500000 33.14333333 35.07000000
##                Max.
## CGAnth   0.69100000
## F1Anth   0.48300000
## Anth201  0.15600000
## Anth103  0.11400000
## CGChl   47.68200000
## F1Chl   59.90500000
## Chl201  59.75700000
## Chl103  58.84300000
## CGFlav   2.45800000
## F1Flav   2.86100000
## Flav201  2.13800000
## Flav103  2.11100000
## CGNBI   22.66000000
## F1NBI   49.17000000
## NBI201  40.97000000
## NBI103  44.88000000

Boxplot comparing medians and measurement distributions

#Boxplot Anthocyanin Allocation
AAnth <- ggplot(Data) + aes(x = Allocation, y = Anth) + geom_boxplot() +geom_hline(yintercept = mean(Data$Anth), color = "purple4") + ylab("Anthocyanin Measure") + xlab("Allocation") + ggtitle("Anthocyanin of Tree Allocation") + theme_classic()+stat_summary(fun.y=mean, colour="darkred", geom="point", shape=18, size=3)

#Boxplot Chlorophyll Allocation
AChl <- ggplot(Data) + aes(x = Allocation, y = Chl) + geom_boxplot() +geom_hline(yintercept = mean(Data$Chl), color = "purple4") + ylab("Chlorophyll Measure") + xlab("Allocation") + ggtitle("Chlorophyll of Tree Allocation") + theme_classic()+stat_summary(fun.y=mean, colour="darkred", geom="point", shape=18, size=3)

#Boxplot Flavanol Allocation
AFlav <- ggplot(Data) + aes(x = Allocation, y = Flav) + geom_boxplot() +geom_hline(yintercept = mean(Data$Flav), color = "purple4") + ylab("Flavanol Measure") + xlab("Allocation") + ggtitle("Flavonol of Tree Allocation") + theme_classic()+stat_summary(fun.y=mean, colour="darkred", geom="point", shape=18, size=3)

#Boxplot NBI Allocation
ANBI <- ggplot(Data) + aes(x = Allocation, y = NBI) + geom_boxplot() +geom_hline(yintercept = mean(Data$NBI), color = "purple4") + ylab("NBI Measure") + xlab("Allocation") + ggtitle("NBI of Tree Allocation") + theme_classic()+stat_summary(fun.y=mean, colour="darkred", geom="point", shape=18, size=3)

#Display Plots
grid.arrange(AAnth,AChl,AFlav,ANBI, ncol=2)

ANOVAs
#Chlorophyll
mod5 <-lm(Data$Chl ~ Data$Allocation)
anova(mod5)
## Analysis of Variance Table
## 
## Response: Data$Chl
##                   Df Sum Sq Mean Sq F value    Pr(>F)    
## Data$Allocation    3   6742 2247.21  9.9149 1.748e-06 ***
## Residuals       1821 412728  226.65                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#Nitrogen
mod6 <-lm(Data$NBI ~ Data$Allocation)
anova(mod5)
## Analysis of Variance Table
## 
## Response: Data$Chl
##                   Df Sum Sq Mean Sq F value    Pr(>F)    
## Data$Allocation    3   6742 2247.21  9.9149 1.748e-06 ***
## Residuals       1821 412728  226.65                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#Flavanol
mod7 <-lm(Data$Flav ~ Data$Allocation)
anova(mod7)
## Analysis of Variance Table
## 
## Response: Data$Flav
##                   Df  Sum Sq Mean Sq F value    Pr(>F)    
## Data$Allocation    3   2.713 0.90423  14.755 1.708e-09 ***
## Residuals       1821 111.598 0.06128                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#Anthocyanin
mod8 <-lm(Data$Anth ~ Data$Allocation)
anova(mod8)
## Analysis of Variance Table
## 
## Response: Data$Anth
##                   Df Sum Sq  Mean Sq F value    Pr(>F)    
## Data$Allocation    3 0.3040 0.101343  47.474 < 2.2e-16 ***
## Residuals       1821 3.8873 0.002135                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

F1 Samples (Approx 3 years) appear more similar to that of the Crimson Glory plant than either/both parents - this is possibly due to age effects as CG is likely more similar in this regard being shorter (No age confirmed). ANOVAs incidate there is significant differences between the allocations - this is to be expected.

Parent Tree Analysis

#Binding parents into single dataset
Parent <- rbind(EC201,EC103)
Parent$No = c(1:22)
#Summary Stats
Parent2 = group_by(Parent, Tree.ID)
Parent2 = summarise(Parent2,Anth = mean(Anth, na.rm = T),Chl = mean(Chl, na.rm = T), Flav = mean(Flav, na.rm = T), NBI = mean(NBI, na.rm = T))
head(Parent2)
## # A tibble: 2 x 5
##   Tree.ID   Anth   Chl  Flav   NBI
##   <fct>    <dbl> <dbl> <dbl> <dbl>
## 1 EC103   0.0588  52.6  1.60  33.1
## 2 EC201   0.0779  52.6  1.71  31.7
#Anthocyanin
ParAnth = summary(Parent$Anth)
#Chlorophyll
ParChl = summary(Parent$Chl)
#Flavonol
ParFlav = summary(Parent$Flav)
#NBI
ParNBI = summary(Parent$NBI)

as.table(rbind(ParAnth,ParChl,ParFlav,ParNBI))
##              Min.   1st Qu.    Median      Mean   3rd Qu.      Max.
## ParAnth  0.004000  0.048250  0.058000  0.067500  0.086750  0.156000
## ParChl  30.722000 50.356750 55.739000 52.598909 57.794750 59.757000
## ParFlav  1.293000  1.475000  1.594000  1.650273  1.821000  2.138000
## ParNBI  14.370000 29.872500 34.025000 32.481818 35.892500 44.880000
#Anthocyanin Parent Plots
#Boxplot
plot5<- ggplot(Parent) + aes(x = Allocation, y = Anth ) + geom_boxplot() + ylab("Anthocyanin Measure") + xlab("Parent") + ggtitle("Anthocyanin of Parent Samples")+stat_summary(fun.y=mean, colour="darkred", geom="point", shape=18, size=3)
#Scatter
plot51<-ggplot(Parent) +aes(x = No, y = Anth) + geom_point() +geom_hline(yintercept = mean(Parent$Anth))+ ylab("Anthocyanin Measure") + xlab("Measure") + ggtitle("Anthocyanin of Parent Samples")
#Histogram
plot52 <-ggplot(Parent) +aes(Anth) +geom_histogram() + stat_bin(binwidth = 0.001) + ylab("Frequency") + xlab("Anthocyanin Measure") + ggtitle("Anthocyanin of Parent Samples")
grid.arrange(plot5,plot51,plot52)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Anthocyanin T.test
t.test(Parent$Anth ~ Parent$Tree.ID)
## 
##  Welch Two Sample t-test
## 
## data:  Parent$Anth by Parent$Tree.ID
## t = -1.2283, df = 19.107, p-value = 0.2342
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -0.05154327  0.01340994
## sample estimates:
## mean in group EC103 mean in group EC201 
##          0.05883333          0.07790000
#Chlorophyll Parent Plots
#Boxplot
plot6<- ggplot(Parent) + aes(x = Allocation, y = Chl ) + geom_boxplot() + ylab("Chlorophyll Measure") + xlab("Parent") + ggtitle("Chlorophyll of Parent Samples")+stat_summary(fun.y=mean, colour="darkred", geom="point", shape=18, size=3)
#Scatter
plot61<-ggplot(Parent) +aes(x = No, y = Chl) + geom_point() +geom_hline(yintercept = mean(Parent$Chl))+ ylab("Chlorophyll Measure") + xlab("Measure") + ggtitle("Chlorophyll of Parent Samples")
#Histogram
plot62 <-ggplot(Parent) +aes(Chl) +geom_histogram() + stat_bin(binwidth = 2) + ylab("Frequency") + xlab("Chlorophyll Measure") + ggtitle("Chlorophyll of Parent Samples")
grid.arrange(plot6,plot61,plot62)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Chlorophyll T.test
t.test(Parent$Chl ~ Parent$Tree.ID)
## 
##  Welch Two Sample t-test
## 
## data:  Parent$Chl by Parent$Tree.ID
## t = -0.003842, df = 17.804, p-value = 0.997
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -7.830954  7.802387
## sample estimates:
## mean in group EC103 mean in group EC201 
##            52.59242            52.60670
#Flavonol Parent Plots
#Boxplot
plot7<- ggplot(Parent) + aes(x = Allocation, y = Flav ) + geom_boxplot() + ylab("Flavonol Measure") + xlab("Parent") + ggtitle("Flavonol of Parent Samples")+stat_summary(fun.y=mean, colour="darkred", geom="point", shape=18, size=3)
#Scatter
plot71<-ggplot(Parent) +aes(x = No, y = Flav) + geom_point() +geom_hline(yintercept = mean(Parent$Flav))+ ylab("Flavonol Measure") + xlab("Measure") + ggtitle("Flavonol of Parent Samples")
#Histogram
plot72 <-ggplot(Parent) +aes(Flav) +geom_histogram() + stat_bin(binwidth = 0.0001) + ylab("Frequency") + xlab("Flavonol Measure") + ggtitle("Flavonol of Parent Samples")
grid.arrange(plot7,plot71,plot72)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Anthocyanin T.test
t.test(Parent$Flav ~ Parent$Tree.ID)
## 
##  Welch Two Sample t-test
## 
## data:  Parent$Flav by Parent$Tree.ID
## t = -1.141, df = 19.612, p-value = 0.2676
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -0.31409194  0.09215861
## sample estimates:
## mean in group EC103 mean in group EC201 
##            1.599833            1.710800
#NBI Parent Plots
#Boxplot
plot8<- ggplot(Parent) + aes(x = Allocation, y = NBI ) + geom_boxplot() + ylab("NBI Measure") + xlab("Parent") + ggtitle("NBI of Parent Samples")+stat_summary(fun.y=mean, colour="darkred", geom="point", shape=18, size=3)
#Scatter
plot81<-ggplot(Parent) +aes(x = No, y = NBI) + geom_point() +geom_hline(yintercept = mean(Parent$NBI))+ ylab("NBI Measure") + xlab("Measure") + ggtitle("NBI of Parent Samples")
#Histogram
plot82 <-ggplot(Parent) +aes(NBI) +geom_histogram() + stat_bin(binwidth = 2) + ylab("Frequency") + xlab("NBI Measure") + ggtitle("NBI of Parent Samples")
grid.arrange(plot8,plot81,plot82)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Anthocyanin T.test
t.test(Parent$NBI ~ Parent$Tree.ID)
## 
##  Welch Two Sample t-test
## 
## data:  Parent$NBI by Parent$Tree.ID
## t = 0.47573, df = 14.7, p-value = 0.6413
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -5.076667  7.987334
## sample estimates:
## mean in group EC103 mean in group EC201 
##            33.14333            31.68800

Interesting sample pattern here, Chl and NBI start low and work high, Flav does the opposite. Maybe accuracy of measurements?

Parental Cross Analysis (F1 Generation)

Investigating measures across the F1 cross population

#Grouping F1 Measures By Tree
F12 <-group_by(F1,Tree.ID)
F12 <- summarise(F12,Chl = mean(Chl,na.rm=T), NBI = mean(NBI, ra.nm = T), Anth = mean(Anth, na.rm=T),Flav = mean(Flav,na.rm =T))
head(F12)
## # A tibble: 6 x 5
##   Tree.ID   Chl   NBI   Anth  Flav
##   <fct>   <dbl> <dbl>  <dbl> <dbl>
## 1 IN4BT    34.5  16.2 0.110   2.17
## 2 IN4BV    27.5  13.8 0.121   2.01
## 3 IN4BW    24.8  12.5 0.127   1.99
## 4 IN4BX    37.6  18.4 0.0905  2.05
## 5 IN4BY    24.2  12.5 0.110   1.98
## 6 IN4BZ    32.0  15.6 0.106   2.09
#Boxplot Anthocyanin F1
FAnth <- ggplot(F1) + aes(x = Tree.ID, y = Anth) + geom_boxplot() +geom_hline(yintercept = mean(F1$Anth), color = "purple4") + ylab("Anthocyanin Measure") + xlab("Tree ID") + ggtitle("Anthocyanin of F1 population") + theme_classic();FAnth

#Mean Measures Anthocyanin
FSAnth <- ggplot(F12) + aes(x = Tree.ID, y = Anth) + geom_point() +geom_hline(yintercept = mean(F12$Anth)) + ylab("Anthocyanin Measure") + xlab("Tree ID") + ggtitle("Anthocyanin of F1 population")
#Histogram of Anthocyanin Distribution
FHAnth <- ggplot(F1) + aes(x = Anth) + geom_histogram()
#Display Anthocyanin Plots
grid.arrange(FSAnth,FHAnth)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Anthocyanin ANOVA
FAA = lm(F1$Anth ~ F1$Tree.ID)
anova(FAA)
## Analysis of Variance Table
## 
## Response: F1$Anth
##              Df Sum Sq   Mean Sq F value    Pr(>F)    
## F1$Tree.ID  158 0.5253 0.0033247  1.8093 2.001e-08 ***
## Residuals  1633 3.0007 0.0018376                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#Boxplot Chlorophyll F1
FChl <- ggplot(F1) + aes(x = Tree.ID, y = Chl) + geom_boxplot() +geom_hline(yintercept = mean(F1$Chl), color = "purple4") + ylab("Chlorophyll Measure") + xlab("Tree ID") + ggtitle("Chlorophyll of F1 Population") + theme_classic();FChl

#Mean Measures Chlorophyll
FSChl <- ggplot(F12) + aes(x = Tree.ID, y = Chl) + geom_point() +geom_hline(yintercept = mean(F12$Chl)) + ylab("Chlorophyll Measure") + xlab("Tree ID") + ggtitle("Chlorophyll of F1 population")
#Histogram of Chlorophyll
FHChl <- ggplot(F1) + aes(x = Chl) + geom_histogram()
#Display Chlorophyll Plots
grid.arrange(FSChl, FHChl)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Chlorophyll ANOVA
FCA = lm(F1$Chl ~ F1$Tree.ID)
anova(FCA)
## Analysis of Variance Table
## 
## Response: F1$Chl
##              Df Sum Sq Mean Sq F value   Pr(>F)   
## F1$Tree.ID  158  47130  298.29  1.3395 0.004474 **
## Residuals  1633 363634  222.68                    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#Boxplot Flavanol F1
FFlav <- ggplot(F1) + aes(x = Tree.ID, y = Flav) + geom_boxplot() +geom_hline(yintercept = mean(F1$Flav), color = "purple4") + ylab("Flavanol Measure") + xlab("Tree ID") + ggtitle("Flavonol of F1 Population") + theme_classic();FFlav

#Mean Measures Flavonol
FSFlav <- ggplot(F12) + aes(x = Tree.ID, y = Flav) + geom_point() +geom_hline(yintercept = mean(F12$Flav)) + ylab("Flavonol Measure") + xlab("Tree ID") + ggtitle("Flavonol of F1 population")
#Histogram of Flavonol
FHFlav <- ggplot(F1) + aes(x = Flav) + geom_histogram()
#Display Chlorophyll Plots
grid.arrange(FSFlav, FHFlav)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Flavonol ANOVA
FFA = lm(F1$Flav ~ F1$Tree.ID)
anova(FFA)
## Analysis of Variance Table
## 
## Response: F1$Flav
##              Df Sum Sq Mean Sq F value    Pr(>F)    
## F1$Tree.ID  158 25.118 0.15897  3.0531 < 2.2e-16 ***
## Residuals  1633 85.030 0.05207                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#Boxplot NBI F1
FNBI <- ggplot(F1) + aes(x = Tree.ID, y = NBI) + geom_boxplot() +geom_hline(yintercept = mean(F1$NBI), color = "purple4") + ylab("NBI Measure") + xlab("Tree ID") + ggtitle("NBI of F1 Population") + theme_classic();FNBI

#Mean Measures NBI
FSNBI <- ggplot(F12) + aes(x = Tree.ID, y = NBI) + geom_point() +geom_hline(yintercept = mean(F12$NBI)) + ylab("NBI Measure") + xlab("Tree ID") + ggtitle("NBI of F1 population")
#Histogram of NBI
FHNBI <- ggplot(F1) + aes(x = NBI) + geom_histogram()
#Display NBI Plots
grid.arrange(FSNBI, FHNBI)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#NBI ANOVA
FNA = lm(F1$NBI ~ F1$Tree.ID)
anova(FNA)
## Analysis of Variance Table
## 
## Response: F1$NBI
##              Df Sum Sq Mean Sq F value    Pr(>F)    
## F1$Tree.ID  158  16378 103.661  1.5147 8.435e-05 ***
## Residuals  1633 111761  68.439                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
Msr = group_by(F1, Tree.ID, measure)
Msr = summarise(Msr, Anth = mean(Anth), Flav = mean(Flav), Chl = mean(Chl),NBI = mean(NBI))

#Select Random Column
#sample(1:4,10, replace = T)
#[1] 3 3 2 1 1 4 1 2 4 3

#Select Random Row
#sample(1:50,10, replace = T)
#[1] 34 37 44 20 44 47  9 19 22 40

#Col 3, Row 34 = IN4G5
IN4G5 = Msr[c(1432:1442),]
AnthG5 <-ggplot(IN4G5) + aes(x = measure, y = Anth) + geom_point() + geom_hline(yintercept = mean(IN4G5$Anth))
ChlG5 <-ggplot(IN4G5) + aes(x = measure, y = Chl) + geom_point() + geom_hline(yintercept = mean(IN4G5$Chl))
FlavG5 <-ggplot(IN4G5) + aes(x = measure, y = Flav) + geom_point() + geom_hline(yintercept = mean(IN4G5$Flav))
NBIG5 <-ggplot(IN4G5) + aes(x = measure, y = NBI) + geom_point() + geom_hline(yintercept = mean(IN4G5$NBI))
grid.arrange(AnthG5,ChlG5,FlavG5,NBIG5)

#Col 3, Row 37 = IN4EP
IN4EP = Msr[c(935:945),]
AnthEP <-ggplot(IN4EP) + aes(x = measure, y = Anth) + geom_point() + geom_hline(yintercept = mean(IN4EP$Anth))
ChlEP <-ggplot(IN4EP) + aes(x = measure, y = Chl) + geom_point() + geom_hline(yintercept = mean(IN4EP$Chl))
FlavEP <-ggplot(IN4EP) + aes(x = measure, y = Flav) + geom_point() + geom_hline(yintercept = mean(IN4EP$Flav))
NBIEP <-ggplot(IN4EP) + aes(x = measure, y = NBI) + geom_point() + geom_hline(yintercept = mean(IN4EP$NBI))
grid.arrange(AnthEP,ChlEP,FlavEP,NBIEP)

#Col 2, Row 44 = IN4EY
IN4EY = Msr[c(1012:1022),]
AnthEY <-ggplot(IN4EY) + aes(x = measure, y = Anth) + geom_point() + geom_hline(yintercept = mean(IN4EY$Anth))
ChlEY <-ggplot(IN4EY) + aes(x = measure, y = Chl) + geom_point() + geom_hline(yintercept = mean(IN4EY$Chl))
FlavEY <-ggplot(IN4EY) + aes(x = measure, y = Flav) + geom_point() + geom_hline(yintercept = mean(IN4EY$Flav))
NBIEY <-ggplot(IN4EY) + aes(x = measure, y = NBI) + geom_point() + geom_hline(yintercept = mean(IN4EY$NBI))
grid.arrange(AnthEY,ChlEY,FlavEY,NBIEY)

#Col 1, Row 20 = IN4CD
IN4CD = Msr[c(204:214),]
AnthCD <-ggplot(IN4CD) + aes(x = measure, y = Anth) + geom_point() + geom_hline(yintercept = mean(IN4CD$Anth))
ChlCD <-ggplot(IN4CD) + aes(x = measure, y = Chl) + geom_point() + geom_hline(yintercept = mean(IN4CD$Chl))
FlavCD <-ggplot(IN4CD) + aes(x = measure, y = Flav) + geom_point() + geom_hline(yintercept = mean(IN4CD$Flav))
NBICD <-ggplot(IN4CD) + aes(x = measure, y = NBI) + geom_point() + geom_hline(yintercept = mean(IN4CD$NBI))
grid.arrange(AnthCD,ChlCD,FlavCD,NBICD)

#Col 1, Row 44 = IN4D8
IN4D8 = Msr[c(463:472),]
AnthD8 <-ggplot(IN4D8) + aes(x = measure, y = Anth) + geom_point() + geom_hline(yintercept = mean(IN4D8$Anth))
ChlD8 <-ggplot(IN4D8) + aes(x = measure, y = Chl) + geom_point() + geom_hline(yintercept = mean(IN4D8$Chl))
FlavD8 <-ggplot(IN4D8) + aes(x = measure, y = Flav) + geom_point() + geom_hline(yintercept = mean(IN4D8$Flav))
NBID8 <-ggplot(IN4D8) + aes(x = measure, y = NBI) + geom_point() + geom_hline(yintercept = mean(IN4D8$NBI))
grid.arrange(AnthD8,ChlD8,FlavD8,NBID8)

#Col4, Row 47 = IN4GV
IN4GV = Msr[c(1645:1654),]
AnthGV <-ggplot(IN4GV) + aes(x = measure, y = Anth) + geom_point() + geom_hline(yintercept = mean(IN4GV$Anth))
ChlGV <-ggplot(IN4GV) + aes(x = measure, y = Chl) + geom_point() + geom_hline(yintercept = mean(IN4GV$Chl))
FlavGV <-ggplot(IN4GV) + aes(x = measure, y = Flav) + geom_point() + geom_hline(yintercept = mean(IN4GV$Flav))
NBIGV <-ggplot(IN4GV) + aes(x = measure, y = NBI) + geom_point() + geom_hline(yintercept = mean(IN4GV$NBI))
grid.arrange(AnthGV,ChlGV,FlavGV,NBIGV)

#Col 1, Row 9 = IN4C2
IN4C2 = Msr[c(93:103),]
AnthC2 <-ggplot(IN4C2) + aes(x = measure, y = Anth) + geom_point() + geom_hline(yintercept = mean(IN4C2$Anth))
ChlC2 <-ggplot(IN4C2) + aes(x = measure, y = Chl) + geom_point() + geom_hline(yintercept = mean(IN4C2$Chl))
FlavC2 <-ggplot(IN4C2) + aes(x = measure, y = Flav) + geom_point() + geom_hline(yintercept = mean(IN4C2$Flav))
NBIC2 <-ggplot(IN4C2) + aes(x = measure, y = NBI) + geom_point() + geom_hline(yintercept = mean(IN4C2$NBI))
grid.arrange(AnthC2,ChlC2,FlavC2,NBIC2)

#Col 2, Row 19 = IN4E3
IN4E3 = Msr[c(740:750),]
AnthE3 <-ggplot(IN4E3) + aes(x = measure, y = Anth) + geom_point() + geom_hline(yintercept = mean(IN4E3$Anth))
ChlE3 <-ggplot(IN4E3) + aes(x = measure, y = Chl) + geom_point() + geom_hline(yintercept = mean(IN4E3$Chl))
FlavE3 <-ggplot(IN4E3) + aes(x = measure, y = Flav) + geom_point() + geom_hline(yintercept = mean(IN4E3$Flav))
NBIE3 <-ggplot(IN4E3) + aes(x = measure, y = NBI) + geom_point() + geom_hline(yintercept = mean(IN4E3$NBI))
grid.arrange(AnthE3,ChlE3,FlavE3,NBIE3)

#Col 4, Row 22 = N/A -> Row 23 = IN4HK
IN4HK = Msr[c(1718:1728),]
AnthHK <-ggplot(IN4HK) + aes(x = measure, y = Anth) + geom_point() + geom_hline(yintercept = mean(IN4HK$Anth))
ChlHK <-ggplot(IN4HK) + aes(x = measure, y = Chl) + geom_point() + geom_hline(yintercept = mean(IN4HK$Chl))
FlavHK <-ggplot(IN4HK) + aes(x = measure, y = Flav) + geom_point() + geom_hline(yintercept = mean(IN4HK$Flav))
NBIHK <-ggplot(IN4HK) + aes(x = measure, y = NBI) + geom_point() + geom_hline(yintercept = mean(IN4HK$NBI))
grid.arrange(AnthHK,ChlHK,FlavHK,NBIHK)

#Col 3, Row 40 = IN4GC
IN4GC = Msr[c(1508:1518),]
AnthGC <-ggplot(IN4GC) + aes(x = measure, y = Anth) + geom_point() + geom_hline(yintercept = mean(IN4GC$Anth))
ChlGC <-ggplot(IN4GC) + aes(x = measure, y = Chl) + geom_point() + geom_hline(yintercept = mean(IN4GC$Chl))
FlavGC <-ggplot(IN4GC) + aes(x = measure, y = Flav) + geom_point() + geom_hline(yintercept = mean(IN4GC$Flav))
NBIGC <-ggplot(IN4GC) + aes(x = measure, y = NBI) + geom_point() + geom_hline(yintercept = mean(IN4GC$NBI))
grid.arrange(AnthGC,ChlGC,FlavGC,NBIGC)

Height Analysis

F13 = group_by(F1,Height)
F13 = summarise(F13, Anth = mean(Anth, na.rm = T), Chl = mean(Chl, na.rm = T), Flav = mean(Flav, na.rm = T), NBI = mean(NBI, na.rm = T))
head(F13)
## # A tibble: 3 x 5
##   Height   Anth   Chl  Flav   NBI
##   <fct>   <dbl> <dbl> <dbl> <dbl>
## 1 H      0.105   33.9  1.93  17.9
## 2 L      0.0968  35.6  1.98  18.5
## 3 M      0.0913  35.3  1.98  18.1
#Boxplot Anthocyanin Height
HAnth <- ggplot(F1) + aes(x = Height, y = Anth) + geom_boxplot() +geom_hline(yintercept = mean(F1$Anth), color = "purple4") + ylab("Anthocyanin Measure") + xlab("Height") + ggtitle("Anthocyanin of F1 population by Height") + theme_classic()+stat_summary(fun.y=mean, colour="darkred", geom="point", shape=18, size=3)

#Boxplot Chlorophyll F1
HChl <- ggplot(F1) + aes(x = Height, y = Chl) + geom_boxplot() +geom_hline(yintercept = mean(F1$Chl), color = "purple4") + ylab("Chlorophyll Measure") + xlab("Height") + ggtitle("Chlorophyll of F1 Population") + theme_classic() +stat_summary(fun.y=mean, colour="darkred", geom="point", shape=18, size=3)

#Boxplot Flavanol F1
HFlav <- ggplot(F1) + aes(x = Height, y = Flav) + geom_boxplot() +geom_hline(yintercept = mean(F1$Flav), color = "purple4") + ylab("Flavanol Measure") + xlab("Height") + ggtitle("Flavonol of F1 Population by Height") + theme_classic() +stat_summary(fun.y=mean, colour="darkred", geom="point", shape=18, size=3)

#Boxplot NBI F1
HNBI <- ggplot(F1) + aes(x = Height, y = NBI) + geom_boxplot() +geom_hline(yintercept = mean(F1$NBI), color = "purple4") + ylab("NBI Measure") + xlab("Height") + ggtitle("NBI of F1 Population") + theme_classic() +stat_summary(fun.y=mean, colour="darkred", geom="point", shape=18, size=3)
grid.arrange(HAnth,HChl,HFlav,HNBI)

#Anthocyanin ANOVA
HAA = lm(F1$Anth ~ F1$Height)
anova(HAA)
## Analysis of Variance Table
## 
## Response: F1$Anth
##             Df Sum Sq   Mean Sq F value    Pr(>F)    
## F1$Height    2 0.0526 0.0262759  13.533 1.468e-06 ***
## Residuals 1789 3.4735 0.0019416                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
summary(HAA)
## 
## Call:
## lm(formula = F1$Anth ~ F1$Height)
## 
## Residuals:
##      Min       1Q   Median       3Q      Max 
## -0.10400 -0.02878 -0.00056  0.02666  0.37800 
## 
## Coefficients:
##              Estimate Std. Error t value Pr(>|t|)    
## (Intercept)  0.105002   0.001973  53.232  < 2e-16 ***
## F1$HeightL  -0.008225   0.002622  -3.137  0.00173 ** 
## F1$HeightM  -0.013666   0.002630  -5.197 2.26e-07 ***
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
## 
## Residual standard error: 0.04406 on 1789 degrees of freedom
## Multiple R-squared:  0.0149, Adjusted R-squared:  0.0138 
## F-statistic: 13.53 on 2 and 1789 DF,  p-value: 1.468e-06
#Chlorophyll ANOVA
HCA = lm(F1$Chl ~ F1$Height)
anova(HCA)
## Analysis of Variance Table
## 
## Response: F1$Chl
##             Df Sum Sq Mean Sq F value Pr(>F)
## F1$Height    2    853  426.30  1.8605 0.1559
## Residuals 1789 409911  229.13
#Flavonol ANOVA
HFA = lm(F1$Flav ~ F1$Height)
anova(HFA)
## Analysis of Variance Table
## 
## Response: F1$Flav
##             Df  Sum Sq Mean Sq F value    Pr(>F)    
## F1$Height    2   0.915 0.45758  7.4943 0.0005739 ***
## Residuals 1789 109.232 0.06106                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#NBI ANOVA
HNA = lm(F1$NBI ~ F1$Height)
anova(HNA)
## Analysis of Variance Table
## 
## Response: F1$NBI
##             Df Sum Sq Mean Sq F value Pr(>F)
## F1$Height    2    100  50.044  0.6992 0.4971
## Residuals 1789 128039  71.570

Row Analysis

#Grouping data by ROw
F1R = group_by(F1,Row)
F1R = summarise(F1R,Chl = mean(Chl,na.rm=T), NBI = mean(NBI, ra.nm = T), Anth = mean(Anth, na.rm=T),Flav = mean(Flav,na.rm =T))
head(F1R)
## # A tibble: 6 x 5
##   Row     Chl   NBI   Anth  Flav
##   <fct> <dbl> <dbl>  <dbl> <dbl>
## 1 1      35.6  17.3 0.107   2.09
## 2 10     30.5  15.7 0.112   1.98
## 3 11     29.9  15.0 0.112   2.04
## 4 12     34.6  18.7 0.0968  1.84
## 5 13     31.3  17.5 0.107   1.86
## 6 14     35.0  17.9 0.0888  1.98
#Comparing Rows with Differing Tree Numbers

#1 Row with 2 Trees
R2 = F1R[c(9),]
#39 Rows with 3 Trees
R3 = F1R[-c(16,26,4,40,36,5,50,42,44,46,9),]
#10  Rows with 4 Trees
R4 = F1R[c(16,26,4,40,36,5,50,42,44,46),]

#Row Sample Summaries
summary(R2)
##       Row         Chl             NBI             Anth       
##  17     :1   Min.   :32.13   Min.   :17.14   Min.   :0.1163  
##  0      :0   1st Qu.:32.13   1st Qu.:17.14   1st Qu.:0.1163  
##  1      :0   Median :32.13   Median :17.14   Median :0.1163  
##  10     :0   Mean   :32.13   Mean   :17.14   Mean   :0.1163  
##  11     :0   3rd Qu.:32.13   3rd Qu.:17.14   3rd Qu.:0.1163  
##  12     :0   Max.   :32.13   Max.   :17.14   Max.   :0.1163  
##  (Other):0                                                   
##       Flav      
##  Min.   :1.887  
##  1st Qu.:1.887  
##  Median :1.887  
##  Mean   :1.887  
##  3rd Qu.:1.887  
##  Max.   :1.887  
## 
summary(R3)
##       Row          Chl             NBI             Anth        
##  1      : 1   Min.   :29.19   Min.   :14.38   Min.   :0.07868  
##  10     : 1   1st Qu.:32.02   1st Qu.:16.44   1st Qu.:0.08535  
##  11     : 1   Median :34.38   Median :17.81   Median :0.09354  
##  14     : 1   Mean   :34.97   Mean   :18.12   Mean   :0.09677  
##  15     : 1   3rd Qu.:37.05   3rd Qu.:19.46   3rd Qu.:0.10917  
##  16     : 1   Max.   :43.89   Max.   :23.16   Max.   :0.12239  
##  (Other):33                                                    
##       Flav      
##  Min.   :1.820  
##  1st Qu.:1.931  
##  Median :1.974  
##  Mean   :1.973  
##  3rd Qu.:2.037  
##  Max.   :2.092  
## 
summary(R4)
##       Row         Chl             NBI             Anth        
##  12     :1   Min.   :31.33   Min.   :16.99   Min.   :0.08183  
##  13     :1   1st Qu.:34.20   1st Qu.:17.74   1st Qu.:0.09326  
##  23     :1   Median :35.08   Median :18.29   Median :0.09747  
##  32     :1   Mean   :35.09   Mean   :18.48   Mean   :0.09702  
##  41     :1   3rd Qu.:36.36   3rd Qu.:19.10   3rd Qu.:0.10265  
##  45     :1   Max.   :38.42   Max.   :20.71   Max.   :0.10721  
##  (Other):4                                                    
##       Flav      
##  Min.   :1.839  
##  1st Qu.:1.898  
##  Median :1.944  
##  Mean   :1.941  
##  3rd Qu.:1.995  
##  Max.   :2.032  
## 
#Checking Variance (R2 discounted due to sample size of 1 row)
var(R3$Anth)
## [1] 0.0001562666
var(R4$Anth)
## [1] 6.039853e-05
var(R3$Chl)
## [1] 14.76679
var(R4$Chl)
## [1] 4.10476
var(R3$Flav)
## [1] 0.004848629
var(R4$Flav)
## [1] 0.004125519
var(R3$NBI)
## [1] 5.602987
var(R4$NBI)
## [1] 1.155704
#T.test 
t.test(R4$Anth, R3$Anth, alternative = "two.sided", var.equal = FALSE)
## 
##  Welch Two Sample t-test
## 
## data:  R4$Anth and R3$Anth
## t = 0.081174, df = 22.551, p-value = 0.936
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -0.006306855  0.006821440
## sample estimates:
##  mean of x  mean of y 
## 0.09702315 0.09676586
t.test(R4$Chl, R3$Chl, alternative = "two.sided", var.equal = FALSE)
## 
##  Welch Two Sample t-test
## 
## data:  R4$Chl and R3$Chl
## t = 0.13309, df = 27.683, p-value = 0.8951
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -1.702350  1.938809
## sample estimates:
## mean of x mean of y 
##  35.08876  34.97053
t.test(R4$Flav, R3$Flav, alternative = "two.sided", var.equal = FALSE)
## 
##  Welch Two Sample t-test
## 
## data:  R4$Flav and R3$Flav
## t = -1.3877, df = 14.921, p-value = 0.1856
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -0.08156262  0.01725693
## sample estimates:
## mean of x mean of y 
##  1.941185  1.973338
t.test(R4$Flav, R3$Flav, alternative = "two.sided", var.equal = FALSE)
## 
##  Welch Two Sample t-test
## 
## data:  R4$Flav and R3$Flav
## t = -1.3877, df = 14.921, p-value = 0.1856
## alternative hypothesis: true difference in means is not equal to 0
## 95 percent confidence interval:
##  -0.08156262  0.01725693
## sample estimates:
## mean of x mean of y 
##  1.941185  1.973338

There appears to be no signficant differences between Rows with 3 trees and rows with 4 trees for any of the measures.

Row Plots and ANOVAs

#Boxplot Anthocyanin Row
RoAnth <- ggplot(F1) + aes(x = Row, y = Anth) + geom_boxplot() +geom_hline(yintercept = mean(F1$Anth), color = "purple4") + ylab("Anthocyanin Measure") + xlab("Row") + ggtitle("Anthocyanin of F1 population by Row") + theme_classic();RoAnth

#Mean Anthocyanin Row Plot
SRoAnth <- ggplot(F1R) + aes(x = as.numeric(Row), y = Anth) + geom_point() +geom_hline(yintercept = mean(F1R$Anth)) + ylab("Anthocyanin Measure") + xlab("Row") + ggtitle("Mean Anthocyanin of F1 population by Row") + theme_classic()
#Histogram of Anthocyanin
RoHAnth <- ggplot(F1R) + aes(x = Anth) + geom_histogram()
#Display plots
grid.arrange(SRoAnth,RoHAnth)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Anthocyanin ANOVA
RoAA = lm(F1$Anth ~ F1$Row)
anova(RoAA)
## Analysis of Variance Table
## 
## Response: F1$Anth
##             Df Sum Sq   Mean Sq F value    Pr(>F)    
## F1$Row      49 0.2346 0.0047878  2.5339 4.261e-08 ***
## Residuals 1742 3.2914 0.0018895                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#Boxplot Chlorophyll F1
RoChl <- ggplot(F1) + aes(x = Row, y = Chl) + geom_boxplot() +geom_hline(yintercept = mean(F1$Chl), color = "purple4") + ylab("Chlorophyll Measure") + xlab("Row") + ggtitle("Chlorophyll of F1 Population by Row") + theme_classic();RoChl

#Mean Chlorophyll Row Plot
SRoChl <- ggplot(F1R) + aes(x = as.numeric(Row), y = Chl) + geom_point() +geom_hline(yintercept = mean(F1R$Chl)) + ylab("Chlorophyll Measure") + xlab("Row") + ggtitle("Mean Chl of F1 population by Row") + theme_classic()
#Histogram of Chlorophyll
RoHChl <- ggplot(F1R) + aes(x = Chl) + geom_histogram()
#Display plots
grid.arrange(SRoChl,RoHChl)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Anthocyanin ANOVA
RoCA = lm(F1$Chl ~ F1$Row)
anova(RoCA)
## Analysis of Variance Table
## 
## Response: F1$Chl
##             Df Sum Sq Mean Sq F value    Pr(>F)    
## F1$Row      49  21394  436.61  1.9533 0.0001046 ***
## Residuals 1742 389370  223.52                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#Boxplot Flavanol F1
RoFlav <- ggplot(F1) + aes(x = Row, y = Flav) + geom_boxplot() +geom_hline(yintercept = mean(F1$Flav), color = "purple4") + ylab("Flavanol Measure") + xlab("Row") + ggtitle("Flavonol of F1 Population by Row") + theme_classic();RoFlav

#Mean Flavonol Row Plot
SRoFlav <- ggplot(F1R) + aes(x = as.numeric(Row), y = Flav) + geom_point() +geom_hline(yintercept = mean(F1R$Flav)) + ylab("Flavonol Measure") + xlab("Row") + ggtitle("Mean Flavonol of F1 population by Row") + theme_classic()
#Histogram of Flavonol
RoHFlav <- ggplot(F1R) + aes(x = Flav) + geom_histogram()
#Display plots
grid.arrange(SRoFlav,RoHFlav)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Anthocyanin ANOVA
RoFA = lm(F1$Flav ~ F1$Row)
anova(RoFA)
## Analysis of Variance Table
## 
## Response: F1$Flav
##             Df  Sum Sq  Mean Sq F value   Pr(>F)    
## F1$Row      49   8.215 0.167646   2.865 3.11e-10 ***
## Residuals 1742 101.933 0.058515                     
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#Boxplot NBI F1
RoNBI <- ggplot(F1) + aes(x = Row, y = NBI) + geom_boxplot() +geom_hline(yintercept = mean(F1$NBI), color = "purple4") + ylab("NBI Measure") + xlab("Row") + ggtitle("NBI of F1 Population by Row") + theme_classic();RoNBI

#Mean NBI Row Plot
SRoNBI <- ggplot(F1R) + aes(x = as.numeric(Row), y = NBI) + geom_point() +geom_hline(yintercept = mean(F1R$NBI)) + ylab("NBI Measure") + xlab("Row") + ggtitle("Mean NBI of F1 population by Row") + theme_classic()
#Histogram of NBI
RoHNBI <- ggplot(F1R) + aes(x = NBI) + geom_histogram()
#Display plots
grid.arrange(SRoNBI,RoHNBI)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Anthocyanin ANOVA
RoNA = lm(F1$NBI ~ F1$Row)
anova(RoNA)
## Analysis of Variance Table
## 
## Response: F1$NBI
##             Df Sum Sq Mean Sq F value   Pr(>F)    
## F1$Row      49   7866 160.527   2.325 8.14e-07 ***
## Residuals 1742 120273  69.043                     
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Column Analysis

#Making Column Values Factors
F1$Column = as.factor(F1$Column)
#Grouping data by Column
F1C = group_by(F1,Column)
F1C = summarise(F1C,Chl = mean(Chl,na.rm=T), NBI = mean(NBI, ra.nm = T), Anth = mean(Anth, na.rm=T),Flav = mean(Flav,na.rm =T))
head(F1C)
## # A tibble: 4 x 5
##   Column   Chl   NBI   Anth  Flav
##   <fct>  <dbl> <dbl>  <dbl> <dbl>
## 1 1       33.8  17.2 0.0945  2.01
## 2 2       35.4  18.6 0.0992  1.94
## 3 3       36.0  18.7 0.0990  1.96
## 4 4       34.3  18.7 0.0912  1.88
#Boxplot Anthocyanin Height
CAnth <- ggplot(F1) + aes(x = Column, y = Anth) + geom_boxplot() +geom_hline(yintercept = mean(F1$Anth)) + ylab("Anthocyanin Measure") + xlab("Column") + ggtitle("Anthocyanin of F1 population by Column") + theme_classic() +stat_summary(fun.y=mean, colour="darkred", geom="point", shape=18, size=3)

#Anthocyanin ANOVA
CAA = lm(F1$Anth ~ F1$Column)
anova(CAA)
## Analysis of Variance Table
## 
## Response: F1$Anth
##             Df Sum Sq   Mean Sq F value Pr(>F)
## F1$Column    3 0.0122 0.0040655  2.0687 0.1024
## Residuals 1788 3.5138 0.0019652
#Boxplot Chlorophyll F1
CChl <- ggplot(F1) + aes(x = Column, y = Chl) + geom_boxplot() +geom_hline(yintercept = mean(F1$Chl), color = "purple4") + ylab("Chlorophyll Measure") + xlab("Column") + ggtitle("Chlorophyll of F1 Population by Column") + theme_classic()+stat_summary(fun.y=mean, colour="darkred", geom="point", shape=18, size=3)

#Chlorophyll ANOVA
CCA = lm(F1$Chl ~ F1$Column)
anova(CCA)
## Analysis of Variance Table
## 
## Response: F1$Chl
##             Df Sum Sq Mean Sq F value  Pr(>F)  
## F1$Column    3   1586  528.66  2.3101 0.07453 .
## Residuals 1788 409178  228.85                  
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#Boxplot Flavanol F1
CFlav <- ggplot(F1) + aes(x = Column, y = Flav) + geom_boxplot() +geom_hline(yintercept = mean(F1$Flav), color = "purple4") + ylab("Flavanol Measure") + xlab("Column") + ggtitle("Flavonol of F1 Population by Column") + theme_classic() +stat_summary(fun.y=mean, colour="darkred", geom="point", shape=18, size=3)

#Flavonol ANOVA
CFA = lm(F1$Flav ~ F1$Column)
anova(CFA)
## Analysis of Variance Table
## 
## Response: F1$Flav
##             Df  Sum Sq Mean Sq F value    Pr(>F)    
## F1$Column    3   2.305 0.76823  12.737 3.087e-08 ***
## Residuals 1788 107.843 0.06031                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#Boxplot NBI F1
CNBI <- ggplot(F1) + aes(x = Column, y = NBI) + geom_boxplot() +geom_hline(yintercept = mean(F1$NBI), color = "purple4") + ylab("NBI Measure") + xlab("Column") + ggtitle("NBI of F1 Population by Column") + theme_classic() +stat_summary(fun.y=mean, colour="darkred", geom="point", shape=18, size=3)
#NBI ANOVA
CNA = lm(F1$NBI ~ F1$Column)
anova(CNA)
## Analysis of Variance Table
## 
## Response: F1$NBI
##             Df Sum Sq Mean Sq F value   Pr(>F)   
## F1$Column    3    821 273.626  3.8427 0.009333 **
## Residuals 1788 127318  71.207                    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
grid.arrange(CAnth,CChl,CFlav,CNBI)

#Anthocyanin
CRHA = lm(F1$Anth ~ F1$Height*F1$Column*F1$Row)
anova(CRHA)
## Analysis of Variance Table
## 
## Response: F1$Anth
##                              Df  Sum Sq   Mean Sq F value    Pr(>F)    
## F1$Height                     2 0.05255 0.0262759 14.9537 3.788e-07 ***
## F1$Column                     3 0.01228 0.0040942  2.3300 0.0727244 .  
## F1$Row                       49 0.23127 0.0047199  2.6861 6.175e-09 ***
## F1$Height:F1$Column           6 0.02055 0.0034247  1.9490 0.0699958 .  
## F1$Height:F1$Row             98 0.20024 0.0020433  1.1628 0.1394606    
## F1$Column:F1$Row            106 0.28473 0.0026861  1.5287 0.0006952 ***
## F1$Height:F1$Column:F1$Row  212 0.41375 0.0019517  1.1107 0.1490621    
## Residuals                  1315 2.31066 0.0017572                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#Remove insignificant variables
CRHA2 = lm(F1$Anth ~ F1$Height*F1$Row)
anova(CRHA2)
## Analysis of Variance Table
## 
## Response: F1$Anth
##                    Df  Sum Sq   Mean Sq F value    Pr(>F)    
## F1$Height           2 0.05255 0.0262759 14.2135 7.586e-07 ***
## F1$Row             49 0.23404 0.0047763  2.5837 2.178e-08 ***
## F1$Height:F1$Row   98 0.20394 0.0020810  1.1257    0.1942    
## Residuals        1642 3.03551 0.0018487                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#Chlorophyll
CRHC = lm(F1$Chl ~ F1$Height*F1$Column*F1$Row)
anova(CRHC)
## Analysis of Variance Table
## 
## Response: F1$Chl
##                              Df Sum Sq Mean Sq F value    Pr(>F)    
## F1$Height                     2    853  426.30  2.0422  0.130152    
## F1$Column                     3   1577  525.57  2.5178  0.056677 .  
## F1$Row                       49  21813  445.16  2.1326 1.244e-05 ***
## F1$Height:F1$Column           6   1282  213.61  1.0233  0.408244    
## F1$Height:F1$Row             98  27033  275.85  1.3215  0.022708 *  
## F1$Column:F1$Row            106  24037  226.77  1.0863  0.265386    
## F1$Height:F1$Column:F1$Row  212  59670  281.46  1.3484  0.001427 ** 
## Residuals                  1315 274499  208.74                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#Remove insignificant variables
CRHC2 = lm(F1$Chl ~ F1$Height*F1$Row)
anova(CRHC2)
## Analysis of Variance Table
## 
## Response: F1$Chl
##                    Df Sum Sq Mean Sq F value   Pr(>F)    
## F1$Height           2    853  426.30  1.9372  0.14444    
## F1$Row             49  21279  434.27  1.9734 8.34e-05 ***
## F1$Height:F1$Row   98  27286  278.43  1.2652  0.04464 *  
## Residuals        1642 361346  220.06                     
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#Further Remove insignificant variables
CRHC3 = lm(F1$Chl ~ F1$Row)
anova(CRHC3)
## Analysis of Variance Table
## 
## Response: F1$Chl
##             Df Sum Sq Mean Sq F value    Pr(>F)    
## F1$Row      49  21394  436.61  1.9533 0.0001046 ***
## Residuals 1742 389370  223.52                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#Flavonol
CRHF = lm(F1$Flav ~ F1$Height*F1$Column*F1$Row)
anova(CRHF)
## Analysis of Variance Table
## 
## Response: F1$Flav
##                              Df Sum Sq Mean Sq F value    Pr(>F)    
## F1$Height                     2  0.915 0.45758  9.3660 9.143e-05 ***
## F1$Column                     3  2.323 0.77447 15.8523 3.957e-10 ***
## F1$Row                       49  8.139 0.16611  3.4001 1.213e-13 ***
## F1$Height:F1$Column           6  1.004 0.16728  3.4240  0.002334 ** 
## F1$Height:F1$Row             98  5.743 0.05860  1.1995  0.096365 .  
## F1$Column:F1$Row            106 14.790 0.13953  2.8559 < 2.2e-16 ***
## F1$Height:F1$Column:F1$Row  212 12.988 0.06126  1.2540  0.012358 *  
## Residuals                  1315 64.245 0.04886                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#NBI
CRHN = lm(F1$NBI ~ F1$Height*F1$Column*F1$Row)
anova(CRHN)
## Analysis of Variance Table
## 
## Response: F1$NBI
##                              Df Sum Sq Mean Sq F value    Pr(>F)    
## F1$Height                     2    100  50.044  0.7850 0.4563286    
## F1$Column                     3    818 272.802  4.2793 0.0051365 ** 
## F1$Row                       49   7816 159.511  2.5021 8.551e-08 ***
## F1$Height:F1$Column           6    792 132.004  2.0707 0.0539220 .  
## F1$Height:F1$Row             98   8175  83.418  1.3085 0.0268723 *  
## F1$Column:F1$Row            106   7765  73.250  1.1490 0.1510179    
## F1$Height:F1$Column:F1$Row  212  18842  88.877  1.3942 0.0004396 ***
## Residuals                  1315  83831  63.750                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#Remove insignificant variables
CRHM2 = lm(F1$NBI ~ F1$Row*F1$Column)
anova(CRHM2)
## Analysis of Variance Table
## 
## Response: F1$NBI
##                    Df Sum Sq Mean Sq F value    Pr(>F)    
## F1$Row             49   7866 160.527  2.3456 6.383e-07 ***
## F1$Column           3    789 263.021  3.8432   0.00934 ** 
## F1$Row:F1$Column  106   7724  72.864  1.0647   0.31303    
## Residuals        1633 111761  68.439                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Controlling for Height,Column,Row effects

Dat = F1
#Total Mean
Dat$ATmean = mean(Dat$Anth)
Dat$CTmean = mean(Dat$Chl)
Dat$FTmean = mean(Dat$Flav)
Dat$NTmean = mean(Dat$NBI)

#Remove Total Mean
Dat$Anth2 = Dat$Anth - Dat$ATmean
Dat$Chl2 = Dat$Chl - Dat$CTmean
Dat$Flav2 = Dat$Flav - Dat$FTmean
Dat$NBI2 = Dat$NBI - Dat$NTmean

#Height Mean
DatH = group_by(Dat, Height)
DatH = summarise(DatH, HAnth = mean(Anth, na.rm = T), HFlav = mean(Flav,na.rm = T))

#Add Height Mean to Data
Dat =merge(Dat, DatH, by.x = "Height")

#Calculate Height Mean Deviation from Total Mean 
Dat$ATH <- Dat$ATmean - Dat$HAnth
Dat$FTH = Dat$FTmean - Dat$HFlav

#Controlling for Height Anth and Flav
Dat$Anth3 = Dat$Anth2 + Dat$ATH
Dat$Flav3 = Dat$Flav2 + Dat$FTH

#Column Mean
DatC = group_by(Dat, Column)
DatC = summarise(DatC, CFlav = mean(Flav,na.rm = T), CNBI = mean(NBI,na.rm = T))

#Add Height Mean to Data
Dat =merge(Dat, DatC, by.x = "Column")

#Calculate Height Mean Deviation from Total Mean 
Dat$FTH = Dat$FTmean - Dat$CFlav
Dat$NTH = Dat$NTmean - Dat$CNBI

#Controlling for Column Flav and NBI
Dat$Flav4 = Dat$Flav3 + Dat$FTH
Dat$NBI3 = Dat$NBI2 + Dat$NTH

#Row Mean
DatR = group_by(Dat, Row)
DatR = summarise(DatR, RAnth = mean(Anth,na.rm = T),RChl = mean(Chl,na.rm = T), RFlav = mean(Flav,na.rm = T), RNBI = mean(NBI,na.rm = T))

#Add Row Mean to Data
Dat =merge(Dat, DatR, by.x = "Row")

#Calculate Row Mean Deviation from Total Mean 
Dat$ATR = Dat$ATmean - Dat$RAnth
Dat$CTR = Dat$CTmean - Dat$RChl
Dat$FTR = Dat$FTmean - Dat$RFlav
Dat$NTR = Dat$NTmean - Dat$RNBI

#Controlling for Row in Anth, Chl, Flav and NBI
Dat$Anth4 = Dat$Anth3 + Dat$ATR
Dat$Chl3 = Dat$Chl2 + Dat$CTR
Dat$Flav5 = Dat$Flav4 + Dat$FTR
Dat$NBI4 = Dat$NBI3 + Dat$NTR

#Anova Checks
AnAnth = lm(Dat$Anth4 ~ Dat$Height*Dat$Column*Dat$Row)
anova(AnAnth)
## Analysis of Variance Table
## 
## Response: Dat$Anth4
##                                 Df  Sum Sq   Mean Sq F value    Pr(>F)    
## Dat$Height                       2 0.00000 0.0000012  0.0007 0.9993236    
## Dat$Column                       3 0.00900 0.0029994  1.7070 0.1637135    
## Dat$Row                         49 0.00061 0.0000125  0.0071 1.0000000    
## Dat$Height:Dat$Column            6 0.02055 0.0034247  1.9490 0.0699958 .  
## Dat$Height:Dat$Row              98 0.20024 0.0020433  1.1628 0.1394606    
## Dat$Column:Dat$Row             106 0.28473 0.0026861  1.5287 0.0006952 ***
## Dat$Height:Dat$Column:Dat$Row  212 0.41375 0.0019517  1.1107 0.1490621    
## Residuals                     1315 2.31066 0.0017572                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
AnChl = lm(Dat$Chl3 ~ Dat$Height*Dat$Column*Dat$Row)
anova(AnChl)
## Analysis of Variance Table
## 
## Response: Dat$Chl3
##                                 Df Sum Sq Mean Sq F value   Pr(>F)   
## Dat$Height                       2    737  368.65  1.7660 0.171413   
## Dat$Column                       3   1956  652.10  3.1239 0.025073 * 
## Dat$Row                         49    155    3.17  0.0152 1.000000   
## Dat$Height:Dat$Column            6   1282  213.61  1.0233 0.408244   
## Dat$Height:Dat$Row              98  27033  275.85  1.3215 0.022708 * 
## Dat$Column:Dat$Row             106  24037  226.77  1.0863 0.265386   
## Dat$Height:Dat$Column:Dat$Row  212  59670  281.46  1.3484 0.001427 **
## Residuals                     1315 274499  208.74                    
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
AnFlav = lm(Dat$Flav5 ~ Dat$Height*Dat$Column*Dat$Row)
anova(AnFlav)
## Analysis of Variance Table
## 
## Response: Dat$Flav5
##                                 Df Sum Sq  Mean Sq F value    Pr(>F)    
## Dat$Height                       2  0.001 0.000491  0.0101  0.989993    
## Dat$Column                       3  0.012 0.004029  0.0825  0.969582    
## Dat$Row                         49  0.215 0.004387  0.0898  1.000000    
## Dat$Height:Dat$Column            6  1.004 0.167281  3.4240  0.002334 ** 
## Dat$Height:Dat$Row              98  5.743 0.058602  1.1995  0.096365 .  
## Dat$Column:Dat$Row             106 14.790 0.139527  2.8559 < 2.2e-16 ***
## Dat$Height:Dat$Column:Dat$Row  212 12.988 0.061264  1.2540  0.012358 *  
## Residuals                     1315 64.245 0.048855                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
AnNBI = lm(Dat$NBI4 ~ Dat$Height*Dat$Column*Dat$Row)
anova(AnNBI)
## Analysis of Variance Table
## 
## Response: Dat$NBI4
##                                 Df Sum Sq Mean Sq F value    Pr(>F)    
## Dat$Height                       2     80  39.936  0.6264 0.5346458    
## Dat$Column                       3     50  16.555  0.2597 0.8544529    
## Dat$Row                         49     22   0.451  0.0071 1.0000000    
## Dat$Height:Dat$Column            6    792 132.004  2.0707 0.0539220 .  
## Dat$Height:Dat$Row              98   8175  83.418  1.3085 0.0268723 *  
## Dat$Column:Dat$Row             106   7765  73.250  1.1490 0.1510179    
## Dat$Height:Dat$Column:Dat$Row  212  18842  88.877  1.3942 0.0004396 ***
## Residuals                     1315  83831  63.750                      
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
#New Heatmaps
#DatHM = group_by(Dat, Column, Row)
#DatHM = summarise(DatHM, Anth = mean(Anth4,na.rm = T), Chl = mean(Chl3,na.rm = T), Flav = mean(Flav5,na.rm = T), NBI = mean(NBI4,na.rm = T))
#write.xlsx(as.data.frame(DatHM), file = "DatHM.xlsx", sheetName ="Dat",col.names = T, row.names = F)

#Anthocyanin Heatmaps
AnthDoc <-read.xlsx("Anth.xlsx", sheetName = "Anth")
heatmaply(AnthDoc, xlab = "Column",ylab ="Row", Rowv = FALSE, Colv = FALSE, main = "Anth Before Correction")
AnthDoc2 <-read.xlsx("DatHM2.xlsx", sheetName = "Anth")
heatmaply(AnthDoc2, xlab = "Column",ylab ="Row", Rowv = FALSE, Colv = FALSE, main = "Anth After Correction")
#Chlorophyll Heatmaps
ChlDoc<-read.xlsx("Chl.xlsx", sheetName = "Chl")
heatmaply(ChlDoc, xlab = "Column",ylab ="Row", Rowv = FALSE, Colv = FALSE, main = "Chl Before Correction")
ChlDoc2 <-read.xlsx("DatHM2.xlsx", sheetName = "Chl")
heatmaply(ChlDoc2, xlab = "Column",ylab ="Row", Rowv = FALSE, Colv = FALSE, main = "Chl After Correction")
#Flavonol Heatmaps
FlavDoc<-read.xlsx("Flav.xlsx", sheetName = "Flav")
heatmaply(FlavDoc, xlab = "Column",ylab ="Row", Rowv = FALSE, Colv = FALSE, main = "Flav Before Correction")
FlavDoc2 <-read.xlsx("DatHM2.xlsx", sheetName = "Flav")
heatmaply(FlavDoc2, xlab = "Column",ylab ="Row", Rowv = FALSE, Colv = FALSE, main = "Flav After Correction")
#NBI Heatmaps
NBIDoc<-read.xlsx("NBI.xlsx", sheetName = "NBI")
heatmaply(NBIDoc, xlab = "Column",ylab ="Row", Rowv = FALSE, Colv = FALSE, main = "NBI Before Correction")
NBIDoc2 <-read.xlsx("DatHM2.xlsx", sheetName = "NBI")
heatmaply(NBIDoc2, xlab = "Column",ylab ="Row", Rowv = FALSE, Colv = FALSE, main = "NBI After Correction")

Looking at controlled data

DatC = group_by(Dat, Tree.ID)
DatC = summarise(DatC, Anth = mean(Anth4,na.rm = T), Chl = mean(Chl3,na.rm = T), Flav = mean(Flav5,na.rm = T), NBI = mean(NBI4,na.rm = T))
#Anthocyanin
CAM = mean(Dat$Anth4)
#Boxplot
Cplot1 <-ggplot(Dat) +aes(x = Tree.ID, y = Anth4) + geom_boxplot()+ ylab("Anth Measure") + xlab("Tree ID") + ggtitle("Anth of F1 after Corrections")+geom_hline(yintercept = CAM)
Cplot11 <-ggplot(DatC) +aes(Anth) +geom_histogram() + stat_bin(binwidth = 0.001) + ylab("Frequency") + xlab("Anth Measure") + ggtitle("Anth of F1 after Corrections")
grid.arrange(Cplot1,Cplot11)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Previous distribution
grid.arrange(FAnth,FHAnth)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Column Compare
Cplot12 <-ggplot(Dat) +aes(x = Column, y = Anth4) + geom_boxplot()+ ylab("Anth Measure") + xlab("Tree ID") + ggtitle("Anth of F1 after Corrections")+geom_hline(yintercept = CAM);Cplot12

#Previous distribution comparison
grid.arrange(CAnth,Cplot12, ncol = 2)

# Row Compare
RoAnth

Cplot13 <-ggplot(Dat) +aes(x = Row, y = Anth4) + geom_boxplot()+ ylab("Anth Measure") + xlab("Row") + ggtitle("Anth of F1 after Corrections")+geom_hline(yintercept = CAM);Cplot13

#Height Compare
Cplot14 <-ggplot(Dat) +aes(x = Height, y = Anth4) + geom_boxplot()+ ylab("Anth Measure") + xlab("Height") + ggtitle("Anth of F1 after Corrections")+geom_hline(yintercept = CAM)
grid.arrange(HAnth,Cplot14,ncol=2)

#Chlorophyll
CCM = mean(Dat$Chl3)
#Boxplot
Cplot2 <-ggplot(Dat) +aes(x = Tree.ID, y = Chl3) + geom_boxplot()+ ylab("Chl Measure") + xlab("Tree ID") + ggtitle("Chl of F1 after Corrections")+geom_hline(yintercept = CCM)
#Histogram
Cplot21 <-ggplot(DatC) +aes(Chl) +geom_histogram() + stat_bin(binwidth = 0.01) + ylab("Frequency") + xlab("Chl Measure") + ggtitle("Chl of F1 after Corrections")
grid.arrange(Cplot2,Cplot21)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Previous Distribution
grid.arrange(FChl,FHChl)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Column Compare
Cplot22 <-ggplot(Dat) +aes(x = Column, y = Chl3) + geom_boxplot()+ ylab("Chl Measure") + xlab("Column") + ggtitle("Chl of F1 after Corrections")+geom_hline(yintercept = CAM)
#Previous distribution
grid.arrange(CChl,Cplot22, ncol = 2)

# Row Compare
RoChl

Cplot23 <-ggplot(Dat) +aes(x = Row, y = Chl3) + geom_boxplot()+ ylab("Chl Measure") + xlab("Row") + ggtitle("Chl of F1 after Corrections")+geom_hline(yintercept = CAM);Cplot23

#Height Compare
Cplot24 <-ggplot(Dat) +aes(x = Height, y = Chl3) + geom_boxplot()+ ylab("Chl Measure") + xlab("Height") + ggtitle("Chl of F1 after Corrections")+geom_hline(yintercept = CCM)
grid.arrange(HChl,Cplot24,ncol=2)

#Flavonol
CFM = mean(Dat$Flav5)
#Boxplot
Cplot3 <-ggplot(Dat) +aes(x = Tree.ID, y = Flav5) + geom_boxplot()+ ylab("Flav Measure") + xlab("Tree ID") + ggtitle("Flav of F1 after Corrections"+geom_hline(yintercept = CFM,color = "purple4"))
#Histogram
Cplot31 <-ggplot(DatC) +aes(Flav) +geom_histogram() + stat_bin(binwidth = 0.001) + ylab("Frequency") + xlab("Flav Measure") + ggtitle("Flav of F1 after Corrections")
grid.arrange(Cplot3,Cplot31)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Previous Distribution
grid.arrange(FFlav,FHFlav)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Column Compare
Cplot32 <-ggplot(Dat) +aes(x = Column, y = Flav5) + geom_boxplot()+ ylab("Flav Measure") + xlab("Column") + ggtitle("Flav of F1 after Corrections")+geom_hline(yintercept = CFM)
#Previous distribution
grid.arrange(CFlav,Cplot32, ncol = 2)

# Row Compare
RoFlav

Cplot33 <-ggplot(Dat) +aes(x = Row, y = Flav5) + geom_boxplot()+ ylab("Flav Measure") + xlab("Row") + ggtitle("Flav of F1 after Corrections")+geom_hline(yintercept = CFM);Cplot33

#Height Compare
Cplot34 <-ggplot(Dat) +aes(x = Height, y = Anth4) + geom_boxplot()+ ylab("Anth Measure") + xlab("Height") + ggtitle("Flav of F1 after Corrections")+geom_hline(yintercept = CFM)
grid.arrange(HFlav,Cplot34,ncol=2)

#NBI
CNM = mean(Dat$NBI4)
#Boxplot
Cplot4 <-ggplot(Dat) +aes(x = Tree.ID, y = NBI4) + geom_boxplot()+ ylab("NBI Measure") + xlab("Tree ID") + ggtitle("NBI of F1 after Corrections")+geom_hline(yintercept = CNM)
#Histogram
Cplot41 <-ggplot(DatC) +aes(NBI) +geom_histogram() + stat_bin(binwidth = 0.001) + ylab("Frequency") + xlab("NBI Measure") + ggtitle("NBI of F1 after Corrections")
grid.arrange(Cplot4,Cplot41)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Previous Distribution
grid.arrange(FNBI,FHNBI)
## `stat_bin()` using `bins = 30`. Pick better value with `binwidth`.

#Column Compare
Cplot42 <-ggplot(Dat) +aes(x = Column, y = NBI4) + geom_boxplot()+ ylab("NBI Measure") + xlab("Column") + ggtitle("NBI of F1 after Corrections")+geom_hline(yintercept = CNM)
#Previous distribution
grid.arrange(CNBI,Cplot42, ncol = 2)

# Row Compare
RoNBI

Cplot43 <-ggplot(Dat) +aes(x = Row, y = NBI4) + geom_boxplot()+ ylab("NBI Measure") + xlab("Row") + ggtitle("NBI of F1 after Corrections")+geom_hline(yintercept = CNM);Cplot43

#Height Compare
Cplot44 <-ggplot(Dat) +aes(x = Height, y = NBI4) + geom_boxplot()+ ylab("NBI Measure") + xlab("Height") + ggtitle("NBI of F1 after Corrections")+geom_hline(yintercept = CNM)
grid.arrange(HNBI,Cplot44,ncol=2)

Correlations Between Measures

Comparing correlations between the 4 dualex measures

plot9 <-ggplot(Dat) + aes(x = Anth4, y = Chl3) + geom_point()
plot10 <-ggplot(Dat) + aes(x = Anth4, y = Flav5) + geom_point()
plot11<-ggplot(Dat) + aes(x = Anth4, y = NBI4) + geom_point()
plot12<-ggplot(Dat) + aes(x = Chl3, y = Flav5) + geom_point()
plot13<-ggplot(Dat) + aes(x = Chl3, y = NBI4) + geom_point()
plot14<-ggplot(Dat) + aes(x = Flav5, y = NBI4) + geom_point()
grid.arrange(plot9,plot10,plot11,plot12,plot13,plot14)

AnthComp = lm(Anth4 ~ Chl3*Flav5*NBI4, data = Dat)
anova(AnthComp)
## Analysis of Variance Table
## 
## Response: Anth4
##                   Df  Sum Sq  Mean Sq  F value    Pr(>F)    
## Chl3               1 0.30271 0.302706 192.4807 < 2.2e-16 ***
## Flav5              1 0.05792 0.057917  36.8277 1.572e-09 ***
## NBI4               1 0.01331 0.013306   8.4608 0.0036737 ** 
## Chl3:Flav5         1 0.02196 0.021955  13.9607 0.0001925 ***
## Chl3:NBI4          1 0.03008 0.030085  19.1299 1.292e-05 ***
## Flav5:NBI4         1 0.00791 0.007905   5.0266 0.0250834 *  
## Chl3:Flav5:NBI4    1 0.00005 0.000055   0.0348 0.8519736    
## Residuals       1784 2.80562 0.001573                       
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
ChlComp = lm(Chl3 ~ Anth4*Flav5*NBI4, data = Dat)
anova(ChlComp)
## Analysis of Variance Table
## 
## Response: Chl3
##                    Df Sum Sq Mean Sq    F value    Pr(>F)    
## Anth4               1  36383   36383 1.9357e+04 < 2.2e-16 ***
## Flav5               1    383     383 2.0354e+02 < 2.2e-16 ***
## NBI4                1 342014  342014 1.8196e+05 < 2.2e-16 ***
## Anth4:Flav5         1    569     569 3.0285e+02 < 2.2e-16 ***
## Anth4:NBI4          1    115     115 6.1416e+01 7.879e-15 ***
## Flav5:NBI4          1   6552    6552 3.4859e+03 < 2.2e-16 ***
## Anth4:Flav5:NBI4    1      1       1 3.2830e-01    0.5667    
## Residuals        1784   3353       2                         
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
FlavComp = lm(Flav5 ~ Anth4*Chl3*NBI4, data = Dat)
anova(FlavComp)
## Analysis of Variance Table
## 
## Response: Flav5
##                   Df Sum Sq Mean Sq   F value    Pr(>F)    
## Anth4              1  2.455   2.455  198.4244 < 2.2e-16 ***
## Chl3               1  0.105   0.105    8.4580  0.003679 ** 
## NBI4               1 73.486  73.486 5940.1653 < 2.2e-16 ***
## Anth4:Chl3         1  0.438   0.438   35.4071 3.213e-09 ***
## Anth4:NBI4         1  0.121   0.121    9.7847  0.001788 ** 
## Chl3:NBI4          1  0.295   0.295   23.8417 1.139e-06 ***
## Anth4:Chl3:NBI4    1  0.028   0.028    2.2925  0.130175    
## Residuals       1784 22.070   0.012                        
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
NBIComp = lm(NBI4 ~ Anth4*Chl3*Flav5, data = Dat)
anova(NBIComp)
## Analysis of Variance Table
## 
## Response: NBI4
##                    Df Sum Sq Mean Sq    F value    Pr(>F)    
## Anth4               1  12013   12013 1.5967e+04 < 2.2e-16 ***
## Chl3                1  95398   95398 1.2679e+05 < 2.2e-16 ***
## Flav5               1   9255    9255 1.2300e+04 < 2.2e-16 ***
## Anth4:Chl3          1     36      36 4.7348e+01 8.199e-12 ***
## Anth4:Flav5         1    118     118 1.5676e+02 < 2.2e-16 ***
## Chl3:Flav5          1   1393    1393 1.8521e+03 < 2.2e-16 ***
## Anth4:Chl3:Flav5    1      1       1 1.6591e+00    0.1979    
## Residuals        1784   1342       1                         
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Flowering

#Plots Flowering vs Non-Flowering
plot39<- ggplot(Dat) + aes(x = Flower, y = Anth4) + geom_boxplot()
plot40<- ggplot(Dat) + aes(x = Flower, y = Chl3) + geom_boxplot()
plot41<- ggplot(Dat) + aes(x = Flower, y = Flav5) + geom_boxplot()
plot42<- ggplot(Dat) + aes(x = Flower, y = NBI4) + geom_boxplot()

grid.arrange(plot39,plot40,plot41,plot42)

AF<- lm(Dat$Anth4 ~ Dat$Flower)
anova(AF)
## Analysis of Variance Table
## 
## Response: Dat$Anth4
##              Df Sum Sq    Mean Sq F value Pr(>F)
## Dat$Flower    1 0.0002 0.00023977  0.1325 0.7159
## Residuals  1790 3.2393 0.00180967
CF<- lm(Dat$Chl3 ~ Dat$Flower)
anova(CF)
## Analysis of Variance Table
## 
## Response: Dat$Chl3
##              Df Sum Sq Mean Sq F value Pr(>F)
## Dat$Flower    1     37  36.997  0.1701 0.6801
## Residuals  1790 389333 217.505
FF<- lm(Dat$Flav5 ~ Dat$Flower)
anova(FF)
## Analysis of Variance Table
## 
## Response: Dat$Flav5
##              Df Sum Sq  Mean Sq F value Pr(>F)  
## Dat$Flower    1  0.248 0.247514  4.4866 0.0343 *
## Residuals  1790 98.750 0.055168                 
## ---
## Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
NF<- lm(Dat$NBI4 ~ Dat$Flower)
anova(NF)
## Analysis of Variance Table
## 
## Response: Dat$NBI4
##              Df Sum Sq Mean Sq F value Pr(>F)
## Dat$Flower    1     12  11.938  0.1788 0.6725
## Residuals  1790 119544  66.784

Looking at Environment and Height

EH = read.xlsx("HeightL.xlsx", sheetName = "Sheet1")
heatmaply(EH,Colv = FALSE, Rowv = FALSE, xlab = "Column",ylab = "Row", main = "Tree Height Across Planting Block")